Journal of Economics and Sustainable Development                                              www.iiste.orgISSN 2222-1700 ...
Journal of Economics and Sustainable Development                                                www.iiste.orgISSN 2222-170...
Journal of Economics and Sustainable Development                                             www.iiste.orgISSN 2222-1700 (...
Journal of Economics and Sustainable Development                                               www.iiste.orgISSN 2222-1700...
Journal of Economics and Sustainable Development                                                  www.iiste.orgISSN 2222-1...
Journal of Economics and Sustainable Development                                               www.iiste.orgISSN 2222-1700...
Journal of Economics and Sustainable Development                                               www.iiste.orgISSN 2222-1700...
Journal of Economics and Sustainable Development                                                       www.iiste.orgISSN 2...
Journal of Economics and Sustainable Development                                                    www.iiste.orgISSN 2222...
Journal of Economics and Sustainable Development                                              www.iiste.orgISSN 2222-1700 ...
Journal of Economics and Sustainable Development                                                www.iiste.orgISSN 2222-170...
Journal of Economics and Sustainable Development                                                www.iiste.orgISSN 2222-170...
Journal of Economics and Sustainable Development                                                 www.iiste.orgISSN 2222-17...
Journal of Economics and Sustainable Development                                     www.iiste.orgISSN 2222-1700 (Paper) I...
Journal of Economics and Sustainable Development                                         www.iiste.orgISSN 2222-1700 (Pape...
Journal of Economics and Sustainable Development                                        www.iiste.orgISSN 2222-1700 (Paper...
Journal of Economics and Sustainable Development                                        www.iiste.orgISSN 2222-1700 (Paper...
Journal of Economics and Sustainable Development                     www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (O...
Journal of Economics and Sustainable Development                           www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2...
Upcoming SlideShare
Loading in …5

21 amassoma ditimi -219-237


Published on

The International Institute for Science, Technology and Education (IISTE) , International Journals Call for papaers:

Published in: Technology, Economy & Finance
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

21 amassoma ditimi -219-237

  1. 1. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011 Components of Government Spending and Economic Growth in Nigeria: An Error Correction Modelling. Amassoma Ditimi (Corresponding Author) Joseph Ayo Babalola University, Ikeji – Arakeji, Osun State, Nigeria. College of social and Management Science, Economics department. Email: Nwosa Philip. The Bells University of Technology, Otta, Abeokuta, Ogun State, Nigeria, Department of Economics. Email: Ajisafe Rufus.Adebayo Obafemi Awolowo University, Ile – Ife, Osun State, Nigeria Email:aoluwatoyin@yahoo.comAbstractThis paper examined the linkage between the components of government spending and economic growth inNigeria. In contrast to existing studies, this study examines the relationship between the components ofgovernment expenditure (that is, agriculture; education; health and transport and communication) andeconomic growth with data spanning from 1970 to 2010. The result of the study showed that expenditureon agriculture had a significant influence on economic growth while expenditure on education, health andtransport and communication had insignificant influence on economic growth. Based on the findings, thisstudy suggests the need for a reversal in declining budgetary allocation to the educational and health sectorin order to provide the sectors with the needed revenue which is necessary in influencing aggregate outputof the economy. In addition, this study recommends the need to redirect the excessive expenditures ofgovernment on its officials in both the house of senate and house of representative to these pivotal sectorsthat is capable of stimulating economic growth of the Nigerian economy. In addition, it is highlyrecommended that the government and relevant stake holders should ensure that funds which are meant fordevelopment of the aforementioned sectors should be properly managed. However, the foregoing can beachieved by increasing funds that are meant for anti – corruption in order to enhance economic growth andsustainable development in Nigeria.Keywords: Public Expenditure, Economic Growth, Education, Inflation, Error correction modeling andBudget Constraint.1.0 IntroductionOver the past three decades, there had been increased contention among development economists as to therelationship between public expenditure and economic growth nexus in Nigeria. While some scholars are of 219
  2. 2. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011the belief that increasing public expenditure enhances economic growth, others are of the view that,increase in government spending have the tendency to slower the performance of an economy in bothdeveloping and developed countries.The foregoing is owing to the problem of determining the actual size of growth that could be encounteredin public expenditure. These among others has preoccupied the attention of many researchers and theoristsin both developed and developing countries For example, (Ezirim & Ofurun, 2003) argued that, the size ofa government spending in a country can be measured in terms of the total public spending expended on it.From the above, it can be deduced that the growth in size can be represented by the index of growth in thesize of government expenditure all things being equal. It is however not wrong to presume that publicexpenditure is a significant variable that can be used to explain economic growth of a country. The questionthat now comes to mind is that what variable can one use to measure the indicators of public expenditureappropriately in developing and developed countries respectively? However, the answer to the abovequestion is not far-fetched from the factor that has been stated earlier.(Abdulliah 2000) and (Al-Yusuf 2003), argues that government performs two major functions, which arenamely; the function of protection and that of provision of public goods respectively. For instance, the roleof protection encompasses the creation of rule of law and enforcement of property right which wouldreduce the risks of criminality, protect life, prosperity; and protection from external aggressions. On theother hand, the government ensures that the provisions of public goods are adequate. These among othersare as follows; good roads education, health and power to mention few. No wonder, some scholars positedthat increase in government expenditure on say, socio-economic and physical infrastructure has thetendency to induce economic growth.Frankly speaking, it should be adduced that government spending on say health and education respectivelycan help enhance productivity of labour by means of increasing growth in form of induced national output.In the same vein, spending on infrastructure such as road, communication, power etc. reduces cost ofproduction of both small and large scale industries which in turn increases private sector investment andprofitability of firms thereby raising economic growth of the nation as observed by (Ranjan 2008; Al-Yusuf& Couray 2009) respectively – who postulated that government expenditure contributes positively toeconomic growth.Particularly in Nigeria government expenditure has continuously increased due to factors such as persistentrise from huge receipt in production and sales of crude oil and the increased demand for public goods suchas; roads, communication, power, education and health. In addition, it is pertinent to pinpoint that there isneed to ensure both internal and external security so as to avoid external invasion in the country.Despite the above premises, the debate has been inconclusive. This is due to the mixed feeling abovedepicting whether or not increasing government spending induces economic growth or not, hence, the needfor this current research. More specifically, the major thrust of this paper is to pin down which specificcomponent of government expenditure significantly impact on economic growth in Nigeria. That is todetermine whether increasing government spending induces economic performance as stipulated by thepast studies or not. 220
  3. 3. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011The objective of this paper therefore is to examine the relationship between the components of governmentexpenditure and economic growth in Nigeria. The remaining part of the paper is structured as follows:Section 2 presents relevant literatures and theoretical framework on the study. Section 3 encompasses themodel specification. Section 4 focuses on the empirical analysis and discussion of results while section 5concludes and proffers policy measures or options for the study or future studies.2.0 Literature Review and Theoretical Framework2.1 Literature ReviewAs a matter of fact, in the underdeveloped countries public expenditure has played an active role inreducing regional disparities through development of social overheads, creation of infrastructure in theform of transport and communication facilities, education and training, growth of capital goods industries,basic and key industries to mention few as opined by (Bhatia 2002) . For instance, the Keynesian modelindicates that during recession a policy of budgetary expansion should be undertaken to increase theaggregate demand in the economy thus boosting the Gross Domestic Product (GDP). It is presumed thatincrease in government spending translates into increased employment in public sector and firms in thebusiness sector. In order words, employment rises, income and profits of suppliers and firms increase, andthis would result in the firms to hire more employees to produce the goods and services ordered by thegovernment. In consonance to the above, the work of (Barro 1990), has stipulated a new perspective inwhich the investigation of the impact of fiscal budgetary expansion via government expenditure canenhance economic growth. No wonder, (Barro & Sula-i-martin 1992) opined that government activityinfluences the direction of economic growth.It should be noticeable that an important way in which public expenditure can accelerate the pace ofeconomic growth is by means of narrowing down the difference between social and private marginalproductivity of certain investments.2.2. Empirical ReviewMany researchers have attempted to investigate the relationship between government expenditure andeconomic growth. For instance, (Ram 1986) study made a rigorous attempt to incorporate a theoreticalbasis for tracing the impacts of government expenditure on economic growth through the use of productionfunction specified for both public and private sectors. The data spanned 115 countries to derive broadgeneralizations for the study. The results revealed that government expenditure has a significant positiveexternality effect on growth particularly in developing countries (LDC) sample but when looked at critically,total government spending has a negative effect on growth. In contrast, the study of (Lin 1994) using asample of 62 countries finds out that investment in non – productive spending has no effect on growth inthe advanced countries but a positive impact in LDC’s.(Junko & Vitali 2008) investigated the impact of government expenditure on economic growth inAzerbaijan due to temporarily oil production boom between (2005-2007) which caused expectationallylarge expenditure increases aimed at improving infrastructure and raising: from the study it was discoveredthat Azerbaijan’s total expenditure increased by a cumulative 160 percent in nominal value from 2005 to2007 with an increase of 41% of non-oil GDP to 74 percent. Similarly, in their research reference were 221
  4. 4. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011made to Nigeria and Saudi Arabia (1970-89) who have also experienced oil boom and increasedgovernment expenditure over the years. After simulating the neo-classical growth model tailored to theAzeri conditions; it was revealed by their result that the evaluated fiscal scenario poses significant risks togrowth sustainability and historical experience which indicates that initial growth performance largelydepends on the efficiency of scale-up expenditure.In line with the above, (Komain & Brahmasrene 2007) examined the association between governmentexpenditures and economic growth in Thailand, by employing the Granger causality test. There resultrevealed that government expenditures and economic growth are not co-integrated. More definitely, theresult indicated a unidirectional relationship as causality runs from government expenditure to growth. Alsothe results depicted that a significant positive effect of government spending on economic growth.Furthermore, (Olugbenga & Owoeye 2007) investigated the relationships between government expenditureand economic growth for a group of 30 OECD countries during the period 1970-2005. The results of theregression showed the existence of a long run relationship between government expenditure and economicgrowth. In addition, the results revealed that there was a unidirectional causality from governmentexpenditure to growth for 16 out of the total countries supplied, thus supporting the Keynesian hypothesis.However, causality was said to run from economic growth to government expenditure in 10 out of thecountries; confirming the Wagner’s law.(Ezirim & Muoghalu 2006) are of the view that the demand effect of government expenditure wouldbecome quite strong even in the presence of balanced budget. The implication of such estrangementbetween demand and supply has the tendency to rear inflationary helixes in the economy as a net result.Similarly; (Baumo 1967), explained the rise in government expenditure in terms of unbalanced growthbetween public and private sectors. This further paved way for dividing the economy into progressiveprivate sector and non-progressive public sector, by explaining that productivity rises only in the privatesector, whereas wage rate rises in both, and as a result public expenditure would rise in the long run. Theoutcome of the above is that as public services are more labour intensive, and hence makes employees tohave no motivation to improve productivity, the increase in public expenditure becomes acceleratory. Nowonder (Rastow 1971) pinpointed that public expenditure is better explained in terms of the changes in thedevelopment level of the country’s economy. For example less developed counties at their foundation ofdevelopment regime higher level of investment in order to create necessary infrastructure for gainfuleconomic growth.(Adamu 2003) posited that special human capacity can be acquired and developed through education,training, health promotion as well as investment in all social services that influences men’s productivecapacities.In the work of (Foister and Henrekson 2001) which studied the relationship between governmentexpenditure and economic growth using a sample of some selected wealthy countries spanning from1970-1995 by employing various econometrics techniques. The author’s finds out that more robust resultsare generated as econometric problems were addressed. Similarly, in India and Saudi Arabia respectively,the effect of government development expenditure was examined as it was discovered that there exist asignificant positive impact of government expenditure on economic growth as opined by (Ranjan 2008) and 222
  5. 5. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011(Al-Yusuf 2000)In support of the above, (Liu-Chin, Hsu & Younis, 2008) examined the relationship between publicexpenditure and economic growth by performing the causality test. From the causality result it was revealedthat total government expenditure causes growth of GDP. While it was contrary or the other hand due to thefact that GDP does not granger cause expansion of government expenditure. It was further revealed fromtheir results that public expenditure raises the US economic growth and as such concludes from the aboveshown by the causality test that the Keynesian hypothesis exerts more influence than the Wagner’s law inUS. In Zealand (Frkin 1988) examined the relationship between government expenditure and economicgrowth by proposing a new framework for New Zealand. The results revealed that higher governmentexpenditure does not upset consumption, but that it instead raises private investment which in turnaccelerates economic growth.It is pertinent to observe from the studies carried out by (Nili & Nafisi 2003; Mohammed 2006) and(Komijani & Memerrejad 2004) in Iran that the contribution of education towards growth in real output hasproven to be higher than the contribution of physical capital. This is because educational enterprises andinstitutions serves as power houses for the production of progressive work force in the country as put by(Saint, 2009). In Nigeria, many researchers have proven in their studies that government spending has acausal relationship with economic growth. For instance studies like that of (Oyinlola 1993) opined thatdefense expenditure has a positive impact on economic growth in Nigeria. Contrary to the above, (Akpan2005) used a disaggregated approach to determine the components of government expenditure whichincludes; capital, recurrent, administrative, economic service, social and community service and transfers)goes a long way to enhance growth and from the results the researchers concluded that there is nosignificant association between most components of government expenditure and economic growth inNigeria.Although this study is not first of its kind using Nigeria data, however it shall go a little further then earlierworks to correctly capture all identified composition of public expenditure during the years under review tobe able to assess the impact of public expenditure on economic growth. This relation is very important fordeveloping countries which Nigeria is inclusive due to increasing government expenditure that is currentlyexperienced. However, this is tended to be associated with rising fiscal deficits suggesting their limitedability to raise deficit revenue to finance higher levels of expenditure. In particular, it is noticeable thatraising deficit tends to retard economic growth in developing countries like Nigeria because of theirinability to check inflation during deficit years. Thus, this study gives a good insight into problems createdby rising government expenditure and how the same impacts on growth.Over and above all, from the array of several literatures it could be deduced that government spending hasboth positive and negative significant effect on economic growth in both developing and developedcountries. But this current study asserts that this assumption cannot be fully believed except if it can bevalidated empirically. More so it is also the thrust of this paper to ascertain the component of governmentspending that enhances growth better and why? However, this current study is among many other studiesthat have investigated the relationship between government spending and economic growth is different.Though, the difference between this and others is that it has included inflation and the level of openness as 223
  6. 6. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011one of the factors that can influence the components of government expenditure among others and growthat large.2.3 Theoretical Underpinning.Over the years, numerous theoretical hypotheses have been offered to explain the size of governmentexpenditure on economic growth. These theories became more substantial since the advent of welfareeconomics which focuses on the expansion of the role of the state especially in the area of providinginfrastructures which has in turn given birth to the theory of public expenditure attracting an increasingattention. This current study focuses on three among the theories of increasing public expenditures. Thefirst theory is the Wagner’s theory, followed by wise man-peacock hypothesis and thirdly the median voterhypothesis.2.3.1 Wagner’s Law of increasing public expenditureThe theory first was associated to a German economist who based his law of increasing state activities onhistorical facts, primarily of Germany. To him, there are inherent tendencies for the activities of differentlayers of a government to increase both intensively and extensively thereby pinpointing that, there is afunctional relationship between the growth of an economy and government activities with the result that thegovernment sector grows faster than the economy in a more specific term, (Wagner 1893) argued thatgovernment spending increases more than proportionately with income, that is, the income elasticity ofdemand for government services is positive and greater than unity through empirical test of this hypothesis.This hypothesis, often tries to find either a positive relationship between government spending and incomeand/or a unidirectional causality running from government spending to income.In particular, Musgrave believes that Wagner was thinking of proportion of public sector in the economy. Insupport of the above (Nitti 1903) concluded that Wagner’s thesis is not only applicable to Germany butthat it can also be applied to other government which differs largely from each other because it has thetendency to induce growth.2.3.2 Wiseman and Peacock HypothesisThe second thesis dealing with the growth of public expenditure was put forth by Wiseman and Peacock intheir study of public expenditure in UK for the period 1890-1955. The main resent of this thesis is thatpublic expenditure does not increase in a smooth and continuous way but that it changes like fashion, this isbecause at times, some social or other disturbance takes place thereby creating a need for increased publicexpenditure which the existing public revenue cannot meet. It should be noted that the earlier insufficiencypressure for public expenditure introduces a constraint on revenue which result to restraining and expansionin public expenditure which in turn results to increased public expenditure and thereby make theinadequacy of the present revenue quite clear to everyone. Hence, the movement from the older level ofexpenditure and taxation to a new and higher level which is known as the Displacement Effect. Hence, thegovernment and the people review the revenue position and the need to find a solution to the importantproblems that have come up and agree to the required adjustments to finance the increased expenditure.Considering the foregoing they now attain a new level of tax tolerance which makes them to be ready totolerate a greater burden of taxation and as a result the general level of expenditure and revenue goes up. In 224
  7. 7. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011this way, the public expenditure and revenue get stabilized at a new level till another disturbance occurs tocause a displacement effect. Thus, each major disturbance leads to the government assents a largerproportion of the total national activity. In other words, there is a concentration effect. The concentrationeffect can also be referred to as the apparent tendency for central government economic activity to growfaster than that of the state and local level government as opined by (Adesoye etal, 2010)2.3.3. The Median Voter HypothesisThe median voter hypothesis states that (under some conditions) government officials choose the level ofgovernment spending selected by the median voter according to (Bowen 1943) and (Black 1958). However,the outcome of such a choice is a demand for public services by the median voter that depends upon suchthings as the median voter’s income and such tax price where this price depends in turn on the voter’s taxshare and the relative unit cost of the public good as given by the technology of public provision.(Borcherding & Deacon 1972) and (Bargstorm & Goodman 1973) are among the first to develop formallyand test empirically the median voter’s model focusing on the spending of local government in the UnitedStates. In a similar view, (Niskaren 1978) extends that such empirical test on the spending behaviour of thefederal government and its aggregate money market behaviour. In light of the above this current studyintegrates the above three reviewed theoretical underpinnings.3.0 Specification and Estimation of The ModelThe model for this study follows the work of (Abu and Abduilahi 2010) in their paper titled “Governmentexpenditure and its impact on economic growth in Nigeria”.In this current study we shall use additional two variables which include inflation and level of openness inthe economy. For instance, from Wagner’s cost over – run argument, it could be deduced that inflationpositively and significantly influences the size of public expenditure; however, this assertion was furtherbuttressed by (Ezirim and Muoghalu 2006). On the other hand (Rodrik 1998) pinpoints that openness hasthe tendency to affect government spending. 3.2 Specification of ModelThis paper investigates the relationship between components of government spending and economy growth.The study would also take a step further to investigate if causation runs between the size of the governmentexpenditure and growth or not. To achieve the above objective, this study would employ Co-integration andError correction modeling. However, the theoretical framework that the study would be based on is theKeynesian and endogenous growth models. This is sequel to the fact that, the Keynesian model states andopines that expansion of government expenditure accelerates economic growth. The growth model is thusspecified as a function of public expenditure. However, inflation and openness were also included tocapture growth in the size of the c0mponents of government spending as shown below.Rgdp= f(opn, inf, torep, tcap, eagr, etac, eexp, hexp) …………….………. (1)Operationalising equation 1 in its linear form gives:Rgdp = β0 + β1opn + β2inf + β3torep + β4tcap + β5eagr + β6etac + β7eexp + β8hexp + u1 ………………………….…………….. (2) 225
  8. 8. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011Assuring a log – linear model the above becomes;Inrgdp = β0 + β1Inopn + β2Ininf + β3Intorep + β4Intcap + β5Ineagr + β6Inetac + β7Ineexp + β8Inhexp + U1 …………….…………………..….. (3)In order to estimate the short-run relationship among the variables, the corresponding error correctionequation is estimated as: p p p p ln rgdpt  0   1 ln rgdpt i   2  ln opnt i   3 ln inft i    4  ln torept i i 1 i 1 i 1 i 1 p p p p  5  ln tcapt i   6  ln eagrt i   7  ln etact i   8 ln eexpt i i 1 i 1 i 1 i 1 p  9  ln hexpt i  ECM t 1  ut .................................................(4) i 1The ECMt-1 is the error correction term. The coefficient of the ECMt-1 measures the speed of adjustmenttoward the long run equilibrium.3.3. Description of VariablesThe variables are measured as follows. Economic growth refers to the changes in real GDP. Real GDP inturn is obtained by dividing GDP at current market price by the consumer price index (CPI). TOREP ismeasured as total recurrent expenditure divided by the CPI. TCAP is captured by the total capitalexpenditure divided by the CPI. EAGR is captured by government expenditure on agriculture divided byCPI. HEXP is measured as government expenditure on health divided by CPI. EEXP is captured bygovernment expenditure on education divided by CPI. ETAC is measured as government expenditure ontransport and communication divided by CPI. OPN is the level of openness in the economy. While, INF isthe inflation rate which is used to measure instability of price in the economy. U refers to the error term.Prior to estimation of the growth model above, standard econometric tests like stationarity test andco-integration test were conducted in order to avoid the generation of spurious regression results.3.4. Sources of DataThis research work employed basically the secondary data sources from central bank of Nigerianpublication i.e. the CBN statistical bulletin from various series which includes; 2006 , 2008 and 2009respectively.. The empirical implementation of the model made use of macroeconomic data covering 39years (1970 – 2010). Our model encompasses the classical, the Keynesian as well as more recent and lessconventional models. The equation specified for the study assisted us to determine the T – Value, F –statistics and Durbin Watson test respectively which were used to test for the significance of the equationspecified. However the co-efficient of determination (R2) was used to measure the rate at which thedependent variable is explained by independent variables.4.0 Empirical Result 226
  9. 9. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 20114.1 Unit Root TestAn empirical analysis of the relationship between economic growth, components of governmentexpenditure and other macroeconomic variables requires appropriate estimation techniques for both thelong run and short run analysis. Thus, this study takes the first step to examine the properties of the timeseries and the extent of co-integration between the variables before proceeding to analysing the long runand short run estimate of equation (3) and (4) respectively. Using the Augmented Dickey-Fuller (ADF) test,table 1 revealed that all the variable were non-stationary at levels at 5 per cent level of significance, thusleading to test at first differences, which revealed that all the variables are stationary at first difference, thatis, integrated of order one I(1).Table 1: Unit Root Test Augmented Dickey-Fuller (ADF) TestVariables Level 1st Difference StatusLtcap -1.6576 -6.5956* I(1)Leagr -0.5900 -9.1668* I(1)Lhexp -0.2125 -5.9558* I(I)Linf -0.0022 -6.4782* I(1)Ltorep 0.0373 -7.8005* I(1)Lrgdp -2.3272 -5.8304* I(1)Opn 2.8091 -70663* I(1)Leexp -0.7949 -7.5094* I(1)Letac 0.3164 -9.1250* I(1)Note: * implies stationarity at one percent level.4. 2. Co-integration EstimateAs evident from the unit root test, all variables are integrated of order one, thus the linear combination ofone or more of these variables might exhibit a long run relationship. In order to capture, the extent ofco-integration among the variables, the multivariate co-integration methodology proposed by (Johansen1990) and Johansen and (Juselius 1991) was utilized. The maximum eigenvalue and the trace test from thistechnique were used to establish the numbers of co-integration vectors and the results are presented in table2 below. Both the results of the trace and maximum eigenvalue tests indicated that, there is oneco-integrating vectors at 5% level of significance. This result suggests that there exist a long runrelationship among the variables in the model.Table 2: Co-integration Test Result 227
  10. 10. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011Trace Test Maximum Eigen value TestNull Alternative Statistics 95% critical Null Alternative Statistics 95% critical values valuesr=0 r≥1 233.426 197.37 r=0 r=1 61.524 58.434r≤1 r≥2 174.902 189.52 r≤1 r=2 49.667 52.363r≤2 r≥3 126.236 127.62 r≤2 r=3 42.701 46.231r≤3 r≥4 83.445 95.734 r≤3 r=4 29.384 40.078r≤4 r≥5 54.061 69.819 r≤4 r=5 20.415 33.877Source: Author’s Computation4.3 Regression Estimate ResultWith respect to the estimated model specified in equation (3), the long run relationship among the variableswas examined. A cursory look at the Ordinary Least Square (OLS) estimate presented in Table 3 revealedthat the explanatory power of the model (R-Squared (89.9 per cent) is very high; this implies that theexplanatory variables in the model explained about 90 per cent of the variations in economic growth whilethe remaining 10 per cent of variations in economic growth is accounted for by other factors not included inthe model. With respect to the variables of interest, it was observed that among the components ofgovernment expenditure, only expenditure on agriculture had a very significant influence on economicgrowth given the value of the t-satistic at 4.9081. This result is in line with our a priori expectation. Incontrast to the above, the contribution of other components of government expenditures (which includeexpenditure on education, health and transport and communication) were observed to be statisticallyinsignificant.Table 3: Long-Run Regression EstimateVariable Coefficient Std. Error t-Statistic Prob.C 1.008429 1.858511 0.542600 0.5912Linf 0.005938 0.005432 1.092999 0.2826Ltcap 0.474740 0.181323 2.618207 0.0134Leagr 0.726164 0.147953 4.908067 0.0000Leexp 0.164863 0.182737 0.902190 0.3737Lhexp 0.098120 0.217217 0.451714 0.6545Letac 0.044041 0.162734 0.270634 0.7884Ltorep 1.255800 0.356925 3.518391 0.0013 228
  11. 11. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011Opn -0.085294 0.027331 -3.120827 0.0038R-Squared = 0.8991; F stat = 35.63(0.000); DW = 1.07916Source: Author’s ComputationTable 4: Residual Stationarity Test Variable Augmented Dickey Fuller (ADF) Test Order of Integration ECM -0.5551 (-3.8856) I(0)Source: Author’s ComputationNOTE: The values in parentheses are t-statistics for the stationarity test for the residual term.As shown in table 2, the null hypothesis stipulates that there is “a random walk” which was rejected at onepercent level of significance, indicating that economic growth and the various components of governmentexpenditure were co-integrated.Following the residual stationarity test, we over parameterized the first differenced form of the variables inequation (4) and used Schwarz Information Criteria to guide parsimonious reduction of the model. Thishelps to identify the main dynamic pattern in the model and to ensure that the dynamics of the model havenot been constrained by inappropriate lag length specification.With respect to the parsimonious regression estimate capturing the short run analysis, it is observed fromtable 4 that there are significant improvement in the parsimonious model of the over parameterized model(see appendix). The Adjust R2, F-stat, and the D.W improved significantly. Overall, the model could beconsidered to be reasonably specified based on its statistical significance and fitness.An examination of the results for the parsimonious error correction model in table 5 showed that theexplanatory power (R2) of the model is relatively high (70%). This implies that the model explained atleast 70% of variations in economic growth. Furthermore, the F-statistics 3.11 (0.008) indicated that themodel fit the data relatively well while the Durbin Watson statistics (1.96) indicates absence ofautocorrelation. The error correction coefficient of the model had the expected negative sign and wassignificant at one per cent.In addition to the above the coefficient of individual variables is examined to determine the relativecontribution of each component of government expenditure to economic growth in Nigeria. The co-efficientof the first lagged value of gross domestic product was positive (0.5565) and significant. The positive effectof the value of previous year’s gross domestic product to current output is inconsistent with a prioriexpectation, implying that a one per cent increase in the first lagged value of gross domestic product iscapable of stimulating current economic growth by 56 per cent. The co-efficient of current inflation rateand the second lagged value of inflation rate were observed to be positive and significant. The positiveeffect of both the current and second lagged value of inflation is consistent with a prior expectation, 229
  12. 12. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011because an increase in commodity price enhances investment and output growth in the current period. Thus,a one per cent increase in current and second lagged value of inflation rate would result in economic growthby 1.0 and 1.4 per cent respectively.The co-efficient of the first and the second lagged value of expenditure on agriculture were observed to bepositive and very significant. The positive effect of both the first and second lagged values of expenditureon agriculture is consistent with a prior expectation, given the immense contribution of the agriculturalsector to economic growth in Nigeria. Thus, a one per cent increase in first and second lagged values ofexpenditure on agriculture would stimulate economic growth by 37.8 and 40.3 per cent respectively. Theco-efficient of the first and the second lagged values of expenditure on education were observed to bepositive and insignificant. The positive effect of both the first and second lagged values of expenditure oneducation is consistent with a priori expectation, but the insignificant effect of the contribution of both thefirst and the second lagged values of expenditure on education can be attributed to the continuous decline inthe budgetary allocation of the government to the educational sector in recent years. Similarly, theco-efficient of current and the second lagged values of expenditure on health were observed to be positiveand insignificant. The positive effect of both the first and second lagged values of expenditure on health isconsistent with a priori expectation, but the insignificant effect of the contribution of both the first and thesecond lagged values of expenditure on health can also be attributed to the continuous decline in thebudgetary allocation of the government to the health sector in recent years.Furthermore, the co-efficient of current and the first lagged value of recurrent expenditures were observedto be positive and insignificant while the coefficient of the second lagged value of the recurrent expenditurewas observed to be positive and significant. Finally, the coefficient of trade openness was observed to benegative and significant. This implies that an increase in the openness of an economy is capable ofretarding the growth of the economy, thus as observed from table 5, an increase in trade openness by oneper cent would bring about a decline in economic growth by 6.9 per cent.Table 5: The Parsimonious Error Correction Regression EstimateVariable Coefficient Std. Error t-Statistic Prob.C 0.258486 0.088699 2.914194 0.0080ecm(-1) -0.534233 0.111152 -4.806340 0.0001Δlrgdp(-1) 0.556488 0.152309 3.653678 0.0014Δlinf 0.010428 0.003190 3.268988 0.0035Δlinf(-2) 0.013863 0.004064 3.411522 0.0025Δleagr(-1) 0.377952 0.094894 3.982902 0.0006Δleagr(-2) 0.402920 0.102078 3.947188 0.0007Δleexp(-1) 0.030239 0.061266 0.493570 0.6298Δleexp(-2) 0.010180 0.009862 1.032292 0.3208Δlhexp 0.255979 0.170045 1.505359 0.1561 230
  13. 13. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011Δlhexp(-2) 0.487454 0.261150 1.866565 0.0847Δltorep 0.367678 0.206337 1.781928 0.0886Δltorep(-1) 0.496243 0.285333 1.739172 0.0960Δltorep(-2) 0.718498 0.261166 2.751116 0.0117Δopn -0.068575 0.026954 -2.544098 0.0185R-Squared = 0.6797; F-Stat = 3.11 (0.0078); DW=1.9636.Source: Author’s Computation5. Conclusion and Policy RecommendationThis paper investigated the relationship between the components of government expenditures (that is,education, agriculture, health and transport and telecommunication) on economic growth in Nigeria for theperiod spanning 1970 to 2010. The results of the long run and short run regression estimate confirmed thatexpenditure on agriculture was the most significant component of government expenditure which impactedon economic growth. However, this contradicted the findings of (Abu, N and, Abdullah, U 2010) whichstates that expenditures on defense and agriculture are not significant in explaining economic growth. Theimpact of the other components (education, health and transport and telecommunication) was observed tobe insignificant in both the long run and the short run. Specifically, the outcomes of the result suggest thatgovernment educational spending has been relatively low which is expected to affect the nation’s level ofhuman capital in the long run. On the other hand, such annual educational spending has the potential to beinfluenced by economic fluctuations in such economy. However, this is consistent with the study of (Akpan2005). Based on these findings, this study offers the following policy recommendations. First, there is theneed for an increase in the budgetary allocation to the agricultural sector and also initiate incentives that canpromote the activities of rural farmers in promoting output growth of the sector. Also, the studyrecommends that the monetary authorities should bridge the widened gap existing between lending rate anddeposit rate to enhance agricultural output in Nigeria. Secondly, the continuous decline in budgetaryallocation to the education and health sector should be reverse as this would act as a catalyst to improveperformance of the sectors and ultimately impact on the aggregate economy. Thirdly, there is the need forthe government to redirect their excessive government revenue in the maintenance of government officialboth in the house of senate and house to representative to these pivotal sectors of the economy. Suchredirection of fund would bring about improve performance of the sectors. 231
  14. 14. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011ReferencesAbdulahi, H. (2000) “The relationship between government expenditure and economic growth in Saudi Arabia” Journal of Administrative Science. 12(2):173 -191.Abu, N and Abdulahi (2010): “Government expenditure and economic growth in Nigeria” A disaggregated 232
  15. 15. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011 Analysis. Business and Economic Journal, volume 4 (5), 41 – 52Adesoye etal (2010) “Dynamic analysis of government spending and economic growth in Nigeria.” Journal of management and society, vol. 1, No 2, pp. 27-37, December Edition 2010.Akpan, M.I. (2005). “Government Expenditure and Economic Growth in Nigeria: A Disaggregated Approach” CBN Economic and Financial Review, 43(1).Al – Yousif (2000) “Does government expenditure inhibit or promote economic growth some empirical evidence from Saudi Arabia”. India Economic Journal, 48(2). 34 – 45Al-yousif and Cooray (2009) “Government expenditure, governance and economic growth” Comparative Economic Studies 51 (3) ,401 – 418.Barro R.I. and Sala -1- Martin (1992) “Public finance in models of economic growth” Review of Economic Studies. 59: 645-661.Barro R.J (1991) “Economic growth in a cross – section of countries’’, Quarterly Journal Of Economics; (12) 6, 407 – 444.Barro, R.J. and Sala – I – Martin, X, (1990) “Government spending in a simple model endogenous growth” Journal Of Political Economy 98(5), 103 – 125.Baumol, N and William J. (1967) “Macroeconomics of unbalanced growth” The Anatorny of the Urban crisis. The American economic JournalBlack Duncan (1958) “The Theory Of Committees and Election”s. Cambridge, U.K: Cambridge University Press.Borgston and Goodman (1973) “Private demand for public goods” The American Economic Review. 63(2), 280 --296.Bowen Howard, R. (1943). “The interpretation of voting in the allocation of economic resources, the quality” Journal of Economics 28(1), 27-48.Butchering and Deacon (1972) “The Demand for the Services of Non-Federal Governments” The American Economic Review 62(4): 891-901.Folster, S. and Henrekson, M. (2001) “Growth Effects of Government Expenditure and Taxation in rich countries”. European Economics Review [].Engen and Skinner (1992) “Fiscal policy and economic growth” NBER working paper 4223.Erkin B, 1988. “Government Expenditure and Economic Growth” Reflections on Professor Ram’s Approach, A NewFramework and Some Evidence from New Zealand Time Series Data. Keio Economic Studies, 25(1): 59-66.Ezirim and Ofurun (2003) “Public expenditure growth and Inflation in Developed and developing countries” Nigeria Business and Social Review, Vol. 2 No1 Jan. Pg. 75 -94Ezirm and Muogbalu (2006) “Explaining the size of public expenditure in less developed countries; Theory 233
  16. 16. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011 and empirical evidence from Nigeria ABSU Journal Of Management Sciences, vol 2, No 2 sept, Pp 134- 154.Junko and Vital IMF (2007) “Impact Of Government Expenditure On Growth” The Case Of Azerbaijan International Monetary Fund Working Paper wp/08/115.Komain and Brah Masrene (2007) “The relationship between government expenditure and economic growth in Thailand” Journal of Economic and Economic Education Research.Komijami and Mernernjad (2004), “Importance of human capital and R and D in economic growth of Iran” Journal of Economics volume 31.16 – 23Laudau D, (1983). “Government Expenditure and Economic Growth”: A Cross Country Study. Southern Economic Journal, 49(2): 783-792.Liu –Chin Hsu and Youmis (2008) “The Association between Government Expenditure and Economic Growth” The Granger Causality Test of the US data, 1947-2002. Journal Of Public Budgeting, Accounting and Financial Management 20 (4):439-452.Lin, S. (1994) “Government spending and economic growth”. Journal of Applied Economics, Volume 26. 43 – 51Mohammad .A. (2006) “The role of Education in Iran economic growth far province education” volume 22(4), pp 57- 84.Nili M and S.H. Nafisi (2003) “Relationship between Human Capital and Economic Growth” Quarterly Journal of Economic Research of Iran, 17(2), 32 - 41Niskanen and Willigm A. Detiats, “Government Spending and Inflation”: What is the evidence. Journal Of Monetary Economics, 4(4) 591-602Olugbenga and Owoeye (2007) “Public Expenditure and Economic growth” New evidence room OECD Countries.Oyinlola (1993) “Nigeria national Defense and Economic Development” An impact analysis Scandianawin Journal of Development Alternative 12(3). 123 – 142Ram R and Biswas (1986). “Military expenditures and economic growth in less developed countries” An augmented model further evidence: Economic development and cultural change, vol 34 (January) p 362-372.Ranja and Shanma (2008) “Government expenditure and economic growth” Evidence from India Icfa University, Journal of public finance 6 (3):60-69[].Rodrikand Dani (1998): “Why do more open economic have bigger government” The journal of political economy 106(5) 997 – 1032 Ram, R (1956) Government size and economic growth. A new framework and some evidence from cross section and time sense Data .American economic review, 76, 191 – 203.Rostow (1971) Politics and the Stages of Growth Cambridge University Press | 1971 | ISBN: 0521081971 | 424 pages | djvu | 2 mb 234
  17. 17. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011Saint (2009) Tertiary Education and Economics Growth in sub-saharan Africa. The World Bank report retrieved April 29, 2009 from saint.htm.Wagner and Adolph (1983) Crundlegung des Politician Okonomie Liepzig . Germany: C.F winter. 235
  18. 18. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011APPENDIXParameterized error correction modelVariable Coefficient Std. Error t-Statistic Prob.C 0.322409 0.158763 2.030765 0.0697ecm(-1) -0.569766 0.222828 -2.556980 0.0285Δlrgdp(-1) 0.567417 0.299071 1.897266 0.0870Δlrgdp(-2) 0.080010 0.208067 0.384536 0.7086Δlinf 0.012055 0.005023 2.399991 0.0373Δlinf(-1) 0.002746 0.005108 0.537458 0.6027Δlinf(-2) 0.016177 0.007396 2.187233 0.0536Δltcap 0.194122 0.299698 0.647726 0.5318Δltcap(-1) 0.103572 0.293437 0.352961 0.7314Δltcap(-2) 0.213132 0.245461 0.868292 0.4056Δleagr 0.037537 0.176990 0.212082 0.8363Δleagr(-1) 0.415821 0.290357 1.432102 0.1826Δleagr(-2) 0.399465 0.205907 1.940021 0.0811Δleexp 0.009816 0.196633 0.049919 0.9612Δleexp(-1) 0.218813 0.211173 1.036179 0.3245Δleexp(-2) 0.415746 0.196697 2.113631 0.0607Δlhexp 0.138868 0.254415 0.545835 0.5971Δlhexp(-1) 0.129786 0.317292 0.409042 0.6911Δlhexp(-2) 0.436711 0.252862 1.727075 0.1149Δletac 0.101504 0.151815 0.668605 0.5189Δletac(-1) 0.066926 0.176947 0.378227 0.7132Δletac(-2) 0.039085 0.162217 0.240941 0.8145Δltorep 0.438164 0.438741 0.998684 0.3415Δltorep(-1) 0.518347 0.514179 1.008107 0.3372Δltorep(-2) 0.916383 0.466844 1.962933 0.0781Δopn -0.089374 0.050306 -1.776615 0.1060Δopn(-1) -0.019711 0.054487 -0.361757 0.7251Δopn(-2) 0.020272 0.059065 0.343214 0.7385 236
  19. 19. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.4, 2011R-squared 0.751515 Mean dependent var 0.133409Adjusted R-squared 0.080605 S.D. dependent var 0.348189S.E. of regression 0.333861 Akaike info criterion 0.782500Sum squared resid 1.114633 Schwarz criterion 1.989142Log likelihood 13.13250 Hannan-Quinn criter. 1.211814F-statistic 1.120143 Durbin-Watson stat 1.986966Prob(F-statistic) 0.447989 237